Skip to content

Your first workspace

This page walks you through creating and managing a virtual machine (VM) in your TRE workspace.

Logging in

  1. Open the TRE portal: https://tvstre.uksouth.cloudapp.azure.com/
  2. You will need the Microsoft Authenticator app on your phone to complete sign-in
  3. Under the Workspaces tab you will see any workspaces assigned to you
  4. Click the workspace title to open it, then click the Guacamole Workspace Service title
  5. Click Connect to open the Guacamole home page, which lists your deployed VMs

Creating a VM

  1. Within the Workspace Service, click Create new to create a user resource
  2. Choose either a Windows or Linux VM template
  3. Fill in a name and description for your VM
  4. Choose a Windows/Linux image — use the OUH Data Science VM image as this has all tools pre-installed. The other options are base images with minimal tooling.

  5. Choose a VM size — see the VM sizes guide for guidance on selecting the right size

    GPU VM resizing

    You cannot resize a GPU VM to a non-GPU VM later. Choose carefully.

  6. Enable shared storage to allow files to be shared between VMs in your workspace

  7. Click Create — deployment takes a few minutes
  8. Once deployed, click Connect to open your VM in the browser

Clipboard access

You may see a pop-up asking Guacamole to use your clipboard — allow this to enable copy-paste between your local machine and the VM.

Starting and stopping your VM

When you are not using your VM, stop it to avoid unnecessary costs.

  1. Find your VM in the TRE portal
  2. Click the three dots (⋯) in the top-right of the VM card
  3. Choose Actions → Stop

The card will show VM deallocated once stopped. Follow the same steps to start it again.

Stopped VMs still hold GPU quota

GPU quota is only released when a VM is deleted, not when it is stopped. See the GPU workloads page for guidance.

Deleting a VM

If you no longer need a VM:

  1. Click the three dots (⋯) and choose Disable
  2. Once disabled, click the three dots again and choose Delete

Available tools

The OUH Data Science VM image includes:

  • R and RStudio
  • Python (via Anaconda) and Jupyter Notebooks
  • Visual Studio Code
  • LibreOffice

Additional tools can be requested via the airlock or as an enhancement request via a GitHub issue in your workspace repository.

TRE guidelines

  • Stop VMs when not in use
  • Delete VMs that are no longer needed
  • Always include a detailed justification when creating airlock requests
  • Report bugs via a GitHub issue in your workspace repository, or via Slack, email, or your workspace manager
  • Data is provided by your workspace manager via the shared storage directory — any new data requests must go through the relevant approval process